A Traffic Sign Object Detection Network Based on YOLOv8- EfficientNet
Zhongyan Zeng, Xingxue Shi, Fengyun Li · 2024
Recognition and detection of traffic signs in autonomous driving is a challenging task. With the advancement of deep learning technology., methods based on convolutional neural networks have made significant progress. However., due to the diverse types., small size., and complex backgrounds of traffic signs., traditional detection networks struggle to capture semantic information in traffic sign images. We propose an improved object detection network for traffic sign recognition and detection., based on the YOLOv8 architecture and leveraging EfficientNet's mixed scaling method. Our approach introduces additional feature pyramid levels to enhance detection performance. Experimental results demonstrate that our method outperforms many state-of-the-art approaches in terms of detection accuracy.